I have perhaps meaningless question about Dirac spinors, but I'm at a loss.

The transformation laws for for left-handed and right-handed 2-spinors are $$ \tag 1 \psi_{a} \to \psi_{a}' = N_{a}^{\quad b} \psi_{b} = \left(e^{\frac{1}{2}\omega^{\mu \nu}\sigma_{\mu \nu}}\right)_{a}^{\quad b}\psi_{b}, \quad \psi^{b}{'} = \psi^{a}(N^{-1})_{a}^{\quad b}, $$ $$ \tag 2 \psi_{\dot {a}} \to \psi_{\dot {a}}' = (N^{*})_{\dot {a}}^{\quad \dot {b}} \psi_{\dot {b}} = \left(e^{\frac{1}{2}\omega^{\mu \nu}\tilde {\sigma}_{\mu \nu}}\right)_{\dot {a}}^{\quad \dot {b}}\psi_{\dot {b}}, \quad \psi^{\dot {b}}{'} = \psi^{\dot {a}}(N^{*^{-1}})_{\dot {a}}^{\quad \dot {b}}, $$ where $$ (\sigma_{\mu \nu})_{a}^{\quad b} = -\frac{1}{4}\left(\sigma_{\mu}\tilde {\sigma}_{\nu}-\sigma_{\nu}\tilde {\sigma}_{\mu}\right), \quad (\tilde {\sigma}_{\mu \nu})_{\quad \dot {a}}^{\dot {b}} = -\frac{1}{4}\left(\tilde {\sigma}_{\mu} \sigma_{\nu}- \tilde {\sigma}_{\nu}\sigma_{\mu}\right), $$ $$ (\sigma_{\mu})_{b\dot {b}} = (\hat {E}, \sigma_{i}), \quad (\tilde {\sigma}_{\nu})^{\dot {a} a} = -\varepsilon^{\dot {a}\dot {b}}\varepsilon^{b a} \sigma_{\dot {b} b} = (\hat {E}, -\sigma_{i}). $$ Why do we always take the Dirac spinor as $$ \Psi = \begin{pmatrix} \varphi_{a} \\ \kappa^{\dot {b}} \end{pmatrix}, $$ not as $$ \Psi = \begin{pmatrix} \varphi_{a} \\ \kappa_{\dot {b}} \end{pmatrix}? $$ According to $(1), (2)$ first one transforms as $$ \delta \Psi ' = \frac{1}{2}\omega^{\mu \nu}\begin{pmatrix}\sigma_{\mu \nu} & 0 \\ 0 & -\tilde {\sigma}_{\mu \nu} \end{pmatrix}\Psi , $$ while the second one - as $$ \delta \Psi ' = \frac{1}{2}\omega^{\mu \nu}\begin{pmatrix}\sigma_{\mu \nu} & 0 \\ 0 & \tilde {\sigma}_{\mu \nu} \end{pmatrix}\Psi , $$ so it is more natural than first, because the first one has both covariant and contravariant components, while the second has only covariant (contravariant components).

  • $\begingroup$ Are the indices a b correct in 1 and 2 ? $\endgroup$ – Your Majesty Mar 8 '14 at 21:13
  • $\begingroup$ @LoveLearning : did you ask about the horisontal position of the indices? If yes, I think so. $\endgroup$ – Andrew McAddams Mar 8 '14 at 21:15
  • $\begingroup$ Maybe I'm too tired but you use b as a summation and as an index etc. $\endgroup$ – Your Majesty Mar 8 '14 at 21:17

I think it is convention to write the conjugate Weyl fermion in, \begin{equation} \left( \begin{array}{c} \phi _\alpha \\ \bar{\kappa} ^{\dot{\beta }} \end{array} \right) \end{equation} (it is common to put a bar over the conjugate representation), with a raised index in order to comply with the ${} _{ \dot{\alpha} } ^{ \,\, \dot{\alpha} } $ contraction of spinor indicies. Recall that we write, \begin{equation} \phi \chi \equiv \phi ^\alpha \chi _\alpha , \quad \psi \bar{\chi} \equiv \phi _{\dot{\alpha}} \bar{\chi} ^{\dot{\alpha}} \end{equation} Thus having the particular index structure for the Dirac spinor gives, \begin{align} \bar{ \Psi } \gamma ^\mu \Psi & = \left( \begin{array}{cc} \kappa ^{\beta } &\bar{ \phi} _{\dot{\alpha}}\end{array} \right) \left( \begin{array}{cc} 0 & ( \sigma ^\mu ) _{ \beta \dot{\beta} } \\ ( \bar{\sigma} ^\mu ) ^{ \dot{\alpha } \alpha } & 0 \end{array} \right) \left( \begin{array}{c} \phi _\alpha \\ \bar{ \kappa} ^{\dot{\beta}} \end{array} \right) \\ & = \kappa \sigma ^\mu \bar{\kappa} + \bar{\phi} \bar{\sigma} ^\mu \phi \end{align} where all the dotted indices contract with an "upwards staircase", ${}_{ \dot{\alpha} } ^{ \,\, \dot{\alpha} } $, and undotted with a "downwards staircase", $ {} ^\alpha _{ \,\, \alpha } $.

  • $\begingroup$ @AndrewMcAddams: No problem, sorry I didn't get a chance to read your previous concern. $\endgroup$ – JeffDror Mar 10 '14 at 1:12

I suspect the origin of this might have to do with the bi-spinor notation. Given a four-vector $b_\mu$, one defines the corresponding bi-spinor, $b\!\!/_{\alpha\dot{\beta}}=b_\mu (\sigma^\mu)_{\alpha\dot{\beta}}$. In this convention, bi-spinors have both lower indices (or upper indices if one uses $(\bar{\sigma}^\mu)^{\beta\dot{\alpha}})$. Once such a choice is made, the index structure of $4\times 4$ gamma matrices is fixed leading to what seems a strange choice for the index structure for a Dirac spinor. In order to avoid such details, I usually use a single meta index $A=(\alpha,\dot{\alpha})$ (capital letters) to denote the combination leaving the finer detail only when I need to work explicitly with gamma matrices. I recommend appendix A of the article by M. Sohnius titled "Introducing Supersymmetry" (Physics Reports 128 (1985) 39-204).

  • $\begingroup$ $b_{\mu}\sigma^{\mu}_{a \dot {b}}$ refers to the $\left( \frac{1}{2}, \frac{1}{2} \right)$ representation, have diffenent transformation law and another equation comparing to bispinor rep. Maybe it is impossible to move from the direct sum of $\left(\frac{1}{2}, 0 \right) + \left( 0, \frac{1}{2}\right)$ rep to the 4-vector one (however, it would be possible if rep is $(1, 0)$ or $(0, 1)$). Can you comment it? $\endgroup$ – Andrew McAddams Mar 9 '14 at 1:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.